Referring now to FIG. 1, a description is made of a conventional radio communication system. The illustrated radio communication system comprises a transmitter apparatus 21 having a plurality of antennas and a receiver apparatus 22 having a plurality of antennas. In the example being illustrated, the transmitter apparatus 21 has two antennas while the receiver apparatus 22 has two antennas.
The transmitter apparatus 21 comprises a serial-to-parallel converting apparatus 101, first and second mapping apparatuses 201-1 and 201-2, and first and second transmitter antennas 104-1 and 104-2. Two sets of the mapping apparatuses are employed in correspondence with the transmitter antennas.
Supplied to the transmitter apparatus 21, an input signal din of a bit stream enters the serial-to-parallel converting apparatus 101. The serial-to-parallel converting apparatus 101 converts the input signal din into first and second parallel signals d1 and d2 which correspond to the number of the transmitter antennas.
The first and second parallel signals d1 and d2 enter the first and second mapping apparatuses 201-1 and 201-2, respectively. In the case that, for example, a 4-value PSK is employed as a modulation level, the first and second mapping apparatuses 201-1 and 201-2 allocate 1 symbol for every 2 bits of the first and second parallel signals (d1, d2). The first and second mapping apparatuses 201-1 and 201-2 convert the first and second parallel signals d1 and d2 into first and second transmission signals s1 and s2, respectively.
The first and second transmission signals s1 and s2 are transmitted via the first and second transmitter antennas 104-1 and 104-2 to the receiver apparatus 22.
The receiver apparatus 22 comprises first and second receiver antennas 105-1 and 105-2, a channel estimating apparatus 106, a demodulating apparatus 108, first and second demapping apparatuses 202-1 and 202-2, and a parallel-to-serial converting apparatus 110. Two sets of the demapping apparatuses are employed in correspondence with the transmitter antennas.
Received in the first and second receiver antennas 105-1 and 105-2, first and second received signals r1 and r2 constitute synthesized waves of the first and second transmission signals s1 and s2.
The channel estimating apparatus 106 calculates a channel matrix H by using a signal which is referred to as, for example, a pilot signal and which is known with respect to both the transmitter apparatus 21 and the receiver apparatus 22. The pilot signal is transmitted from the transmitter apparatus 21 to the receiver apparatus 22 independent from the first and second transmission signals s1 and s2. The channel matrix H corresponds to a matrix which is expressed by the below-mentioned formula 1, and this channel matrix H causes the first and second received signals r1 and r2 to be related to the first and second transmission signals s1 and s2.
                                                        r              =                              (                                                                                                    r                        1                                                                                                                                                r                        2                                                                                            )                                                                                        =                                                                    (                                                                                                                        h                            11                                                                                                                                h                            12                                                                                                                                                                            h                            21                                                                                                                                h                            22                                                                                                                )                                    ⁢                                      (                                                                                                                        s                            1                                                                                                                                                                            s                            2                                                                                                                )                                                  +                                  (                                                                                                              n                          1                                                                                                                                                              n                          2                                                                                                      )                                                                                                        =                              Hs                +                n                                                                        (        1        )            It should be noted that symbols n1 and n2 indicate first and second noise signals which are added to the first and second received signals r1 and r2, respectively.
The channel estimating apparatus 106 calculates the channel matrix H based upon the pilot signal transmitted by the transmitter apparatus 21 and this pilot signal received by the receiver apparatus 22 while the pilot signal is known by also the receiver apparatus 22.
The first and second received signals r1 and r2 enter the demodulating apparatus 108. The demodulating apparatus 108 receives the channel matrix H from the channel estimating apparatus 106, and obtains first and second demodulated signals y1 and y2 by using an inverse matrix H−1 of the channel matrix H in accordance with the below-mentioned formula 2.
                                                        y              =                              (                                                                                                    y                        1                                                                                                                                                y                        2                                                                                            )                                                                                        =                                                H                                      -                    1                                                  ⁢                r                                                                                        =                              s                +                                                      H                                          -                      1                                                        ⁢                  n                                                                                                        =                                                (                                                                                                              s                          1                                                                                                                                                              s                          2                                                                                                      )                                +                                                      1                                                                H                                                                              ⁢                                      (                                                                                                                        h                            22                                                                                                                                -                                                          h                              12                                                                                                                                                                                                        -                                                          h                              21                                                                                                                                                            h                            11                                                                                                                )                                    ⁢                                      (                                                                                                                        n                            1                                                                                                                                                                            n                            2                                                                                                                )                                                                                                                          =                                                (                                                                                                              s                          1                                                                                                                                                              s                          2                                                                                                      )                                +                                                      1                                                                                            h                          11                                                ⁢                                                  h                          22                                                                    -                                                                        h                          12                                                ⁢                                                  h                          21                                                                                                      ⁢                                      (                                                                                                                                                                                      h                                22                                                            ⁢                                                              n                                1                                                                                      -                                                                                          h                                12                                                            ⁢                                                              n                                2                                                                                                                                                                                                                                                                                                    -                                                                  h                                  21                                                                                            ⁢                                                              n                                1                                                                                      +                                                                                          h                                11                                                            ⁢                                                              n                                2                                                                                                                                                                          )                                                                                                          (        2        )            
It should also be noted that although matrix elements hnm(n=1, 2; m=1, 2) of the channel matrix H are dynamically varied over time, if such a case can be regarded in which a correlation among the matrix elements is sufficiently low and these matrix elements are independently varied, then it is known that the inverse matrix H−1 is present. When the inverse matrix H−1 is present, the demodulation can be carried out by the above-explained method.
The first and second demodulated signals y1 and y2 enter the first and second demapping apparatuses 202-1 and 202-2, respectively. The first and second demapping apparatuses 202-1 and 202-2 convert the first and second demodulated signals y1 and y2 into first and second reproduced parallel signals “{circumflex over (d)}1” and “{circumflex over (d)}2” of bit streams respectively in accordance with sequential operations which are completely reversed with respect to those of the first and second mapping apparatuses 201-1 and 201-2. That is, for instance, in such a case that a 4-value PSK is employed as a modulation multi-value number, the demodulated signals (y1, y2) of 1 symbol are converted into reproduced parallel signals of 2 bits.
The first and second reproduced parallel signals “{circumflex over (d)}1” and “{circumflex over (d)}2” enter the parallel-to-serial converting apparatus 110. The parallel-to-serial converting apparatus 110 converts the first and second reproduced parallel signals “{circumflex over (d)}1” and “{circumflex over (d)}2” into an output signal dout, which is produced from the receiver apparatus 22.
On the other hand, in the above-explained conventional radio system, signal-to-noise ratios (SNR) “γ1” and “γ2” of the first and second demodulated signals y1 and y2 are expressed by the below-mentioned formula 3 and formula 4, respectively.
                                          γ            1                    =                                                                                                            s                    1                                                                    2                            ⁢                                                                                                                                    h                        11                                            ⁢                                              h                        22                                                              -                                                                  h                        12                                            ⁢                                              h                        21                                                                                                              2                                                                    σ                1                2                            ⁡                              (                                                      h                    22                                    ⁢                                      ❘                    2                                    ⁢                                      +                                                                                                                    h                          12                                                                                            2                                                                      )                                                    ⁢                                                      (        3        )                                          γ          2                =                                                                                              s                  2                                                            2                        ⁢                                                                                                                      h                      11                                        ⁢                                          h                      22                                                        -                                                            h                      12                                        ⁢                                          h                      21                                                                                                  2                                                          σ              2              2                        ⁡                          (                                                h                  21                                ⁢                                  ❘                  2                                ⁢                                  +                                                                                                          h                        11                                                                                    2                                                              )                                                          (        4        )            It should be noted that σ12 and σ22 denote the noise power. In addition, the noise power σ2 may be calculated by using the below-mentioned formula 5 by using K pieces of sampling time, assuming now that a demodulated signal as a function of time is equal to y(t) (suffix letter is omitted, and omission of suffix letter is repeated in below-mentioned symbols).
                              σ          2                =                              1            K                    ⁢                                    ∑                              t                =                1                            K                        ⁢                                          {                                                      y                    ⁡                                          (                      t                      )                                                        -                                                            1                      K                                        ⁢                                                                  ∑                                                                              t                            ′                                                    =                          1                                                K                                            ⁢                                              y                        ⁡                                                  (                                                      t                            ′                                                    )                                                                                                                    }                            2                                                          (        5        )            
The right side of the above-described formula 5 originally corresponds to an amount Indicative of a sum of interference power and noise power. However, in the above-explained conventional radio communication system, the first and second demodulated signals y1 and y2 are demodulated by using the inverse matrix H−1 of the channel matrix H, and thus, an interference signal component is not present in the first and second demodulated signals y1 and y2, as indicated by the above-described formula 2. This reason is given as follows: That is, the first demodulated signal y1 of the above-described formula 2 does not depend upon the second transmission signal s2, and the second demodulated signal y2 does not depend upon the first transmission signal s1. As a consequence, the right side of the above-described formula 5 becomes equal to the noise power σ2.
As previously explained, if both the first and second demodulated signals y1 and y2 are measured, and are applied to the above-described formula 5, then both the first and second noise power σ12 and σ22 may be obtained. However, generally speaking, the noise power σ2 may be substantially determined based upon a noise signal which is produced in an amplifier contained in the receiver antenna 105. As a consequence, in a case that the same type of receiver antennas and the same type of amplifiers are employed in the same environment, it is possible to regard that the first noise power σ12 of the first demodulated signal y1 is equal to the second noise power σ22 of the second demodulated signal y2 (σ12=σ22). In addition, inasmuch as noise power σ2 is not so closely associated with a temporal change, if the noise power σ2 is once calculated, then this calculated noise power may be employed as a known value (σ12=σ22=const.).
In addition, squared temporal averages |s1|2 and |s2|2 of the first and second transmission, signals correspond to the known values in a radio communication system, and further, equal to each other (|s1|2=|s2|2=const.).
When the channel matrix H is calculated by the above-described method in the above-described conventional radio communication system, first and second SNRs γ1 and γ2 of the first and second demodulated signals y1 and y2 can be calculated.
Now, it is assumed that the matrix elements hnm (n=1, 2; m=1, 2) of the channel matrix H, which are dynamically varied, become such values capable of satisfying, for example, |h22|2+|h12|2=2(|h21|2+|h11|2). In this case, the SNR values become γ2=2γ1. Thus, it can be seen that the SNR value of the second demodulated signal y2 is 2 times larger than the SNR value of the first demodulated signal y1. In this case, when the SNR value of the first demodulated signal y1 is equal to a predetermined value, the SNR value of the second demodulated signal y2 becomes excessive, which does not give merit in view of such a technical point that information is effectively transmitted. In order to transmit the information in an effective manner, a transmission parameter such as a modulation level is adaptively controlled, and thus, the excessively large SNR value is utilized so as to increase the transmission speed, which may give merit.
Among the conventional techniques, several techniques have been involved which control transmission parameters based upon physical amounts used to estimate line qualities, that are typified by SNRs or the like. For instance, in radio communication systems arranged by a transmitter apparatus having a single antenna and a receiver apparatus having a single antenna, such techniques for controlling transmission parameters based upon physical amounts such as a code error rate have been disclosed (see Japanese Laid-open Patent Application No. Sho-57-155856, and Japanese Laid-open Patent Application No. 2001-103041). In addition, another technique has been disclosed in which a single receiver apparatus for communicating with a transmitter apparatus having a plurality of antennas which transmit the same signals exclusively selects the antenna of the transmitter apparatus based upon a line quality (see Japanese Laid-open Patent Application No. 2003-78480).
However, in such a radio communication system for transmitting/receiving data in a parallel manner between a transmitter apparatus having a plurality of antennas and a receiver apparatus having a plurality of antennas, a technique capable of controlling a transmission parameter based upon a line quality has not yet been proposed, while the present invention is directed to the above-explained radio communication system.
An object of the present invention is to provide a radio communication system arranged by a transmitter apparatus having a plurality of antennas and a receiver apparatus having a plurality of antennas, which transmits and receives data in a parallel manner and which controls a transmission parameter based upon a line quality.